Autopilot Predict
- 653 installs
- 67k repo stars
- Updated August 4, 2026
- ruvnet/ruflo
autopilot-predict is a Ruflo agent skill that calls Claude-Flow autopilot MCP tools to forecast the optimal next command or workflow step from learned patterns and current task state.
About
autopilot-predict is a ruvnet/ruflo skill for developers who want coding agents to propose the next action instead of waiting for explicit instructions. It calls `mcp__claude-flow__autopilot_predict` for a recommended step, executes directly when confidence exceeds 0.7, otherwise checks `mcp__claude-flow__autopilot_progress` for incomplete tasks and picks the highest-priority item. After work completes, `mcp__claude-flow__autopilot_learn` records outcomes and `mcp__claude-flow__autopilot_history` supplies prior patterns. Reach for autopilot-predict mid-session when task queues are ambiguous, multi-step refactors need sequencing, or you want the agent to self-route based on Ruflo's learned workflow intelligence rather than static checklists alone.
- Predicts the single most effective next agent command or tool call
- Reduces decision fatigue by surfacing high-confidence autopilot suggestions
- Works across research, coding, debugging, deployment and growth tasks
- Returns confidence scores and alternative options with each prediction
- Designed as a reusable capability inside larger agentic workflows
Autopilot Predict by the numbers
- 653 all-time installs (skills.sh)
- +6 installs in the week ending Aug 5, 2026 (Skillselion tracking)
- Ranked #1,500 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/ruvnet/ruflo --skill autopilot-predictAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 653 |
|---|---|
| repo stars | ★ 67k |
| Last updated | August 4, 2026 |
| Repository | ruvnet/ruflo ↗ |
What should the coding agent do next?
Let their coding agent forecast the next optimal command or workflow step before taking action.
Who is it for?
Developers using Ruflo and Claude-Flow MCP who want confidence-gated next-action predictions during multi-step agent sessions.
Skip if: Developers on fixed single-command tasks or repositories without Claude-Flow autopilot MCP servers configured.
When should I use this skill?
The session needs the optimal next workflow step and Claude-Flow autopilot MCP tools are available for predict, progress, learn, and history.
What you get
Executed next-step command, autopilot learn feedback, and updated task progress from Claude-Flow MCP calls.
- Executed next workflow step
- Autopilot learn feedback entry
- Updated task progress snapshot
By the numbers
- Auto-executes autopilot predictions when confidence exceeds 0.7
Files
Predict what to work on next using Ruflo autopilot intelligence:
1. Call mcp__claude-flow__autopilot_predict for the recommended next action 2. If confidence > 0.7, execute the prediction directly 3. If confidence < 0.7, check mcp__claude-flow__autopilot_progress for task breakdown 4. Pick the highest-priority incomplete task 5. After completing work, call mcp__claude-flow__autopilot_learn to update patterns
Learning Pipeline
mcp__claude-flow__autopilot_learn-- discover success patterns from completed tasksmcp__claude-flow__autopilot_history({ query: "KEYWORD" })-- search past completions- Patterns are stored in AgentDB for cross-session recall
Integration with /loop
When running inside a /loop, the predict skill guides each iteration:
- High confidence prediction -> execute immediately
- Low confidence -> fall back to task list priority order
- No tasks remaining -> disable autopilot and exit loop
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10How it compares
Pick autopilot-predict for learned next-step routing inside Ruflo; use static workflow skills when every step is predefined without MCP prediction.
FAQ
When does autopilot-predict execute automatically?
autopilot-predict executes the recommended action directly when `mcp__claude-flow__autopilot_predict` returns confidence above 0.7. Below that threshold, it consults `autopilot_progress` and selects the highest-priority incomplete task instead.
Which MCP tools does autopilot-predict use?
autopilot-predict uses Claude-Flow MCP tools: autopilot_predict, autopilot_progress, autopilot_learn, and autopilot_history. Predict proposes steps, progress breaks down tasks, learn captures outcomes, and history surfaces prior patterns.
What should happen after completing work?
After completing work, autopilot-predict calls `mcp__claude-flow__autopilot_learn` so Ruflo records the outcome and improves future predictions. This closes the loop between execution and subsequent next-action recommendations.